GAST: Geometry-Aware Structure Transformer

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

68 Downloads (Pure)

Abstract

We present GAST, a novel model for realistic building delineation, trained using noisy, imperfect ortho imagery and designed for real-life applications. While most popular methods today rely on some form of semantic segmentation, the core interest is not the building’s interior points but rather the sequence of points surrounding the outer hull, i.e., the most sparse set of points encapsulating the geometry of the building. Our method works end-to-end, removing the need for post-processing, while demonstrating generalization across large geographical differences. We compare our method to state-of-the-art, complementary works and demonstrate that our model outperforms the baselines in a variety of circumstances and across all metrics relating to polygon fidelity. We release our dataset and model checkpoints at www.huggingface.co/datasets/pihalf/ERBD
Original languageEnglish
Title of host publicationProceedings of 2024 IEEE/CVF Winter Conference on Applications of Computer Vision
PublisherIEEE
Publication date2024
Pages776-784
ISBN (Print)979-8-3503-7071-3
ISBN (Electronic)979-8-3503-7028-7
DOIs
Publication statusPublished - 2024
Event2024 IEEE/CVF Winter Conference on Applications of Computer Vision - Waikōloa Beach Marriott Resort, Waikōloa , United States
Duration: 4 Jan 20248 Jan 2024

Conference

Conference2024 IEEE/CVF Winter Conference on Applications of Computer Vision
LocationWaikōloa Beach Marriott Resort
Country/TerritoryUnited States
CityWaikōloa
Period04/01/202408/01/2024

Fingerprint

Dive into the research topics of 'GAST: Geometry-Aware Structure Transformer'. Together they form a unique fingerprint.

Cite this